Skip to content

Latest commit

 

History

History
45 lines (28 loc) · 1.1 KB

package.md

File metadata and controls

45 lines (28 loc) · 1.1 KB

Alternative Installation Methods

This document explains how to install VGGT as a package using different package managers.

Prerequisites

Before installing VGGT as a package, you need to install PyTorch and torchvision. We don't list these as dependencies to avoid CUDA version mismatches. Install them first, with an example as:

# install pytorch 2.3.1 with cuda 12.1
pip install torch==2.3.1 torchvision==0.18.1 --index-url https://download.pytorch.org/whl/cu121

Installation Options

Install with pip

The simplest way to install VGGT is using pip:

pip install -e .

Install and run with pixi

Pixi is a package management tool for creating reproducible environments.

  1. First, download and install pixi
  2. Then run:
pixi run -e python demo_gradio.py

Install and run with uv

uv is a fast Python package installer and resolver.

  1. First, install uv
  2. Then run:
uv run --extra demo demo_gradio.py